Download 13 Interplay Between H2AX and 53BP1 Pathways in DNA Double

13
Interplay Between H2AX
and 53BP1 Pathways in DNA
Double-Strand Break Repair
Response
Fatouros Chronis, BSc,
and Emmy P. Rogakou,
PhD
Summary
One of the most fascinating themes in the biology of double-strand breaks (DSBs)
is that chromatin is emerging as a multifunctional player in the DSB damage response.
The phosphorylation of H2AX on Ser 139, named H2AX, is an early response to the
generation of DNA DSBs and extends along megabase-long domains, both sites of the
lesion, supporting amplification of signal transduction pathways. In parallel, 53BP1
accumulates on damaged chromatin to interface between methylated histone residues
and proteins that belong to the signal-transduction pathways, mediating cell-cycle
arrest or apoptosis. Interestingly, the two pathways crosstalk at the chromatin level.
Key Words: H2AX; 53BP1; double-strand breaks; DNA repair; chromatin.
1. GENERAL CHARACTERISTICS OF DOUBLE-STRAND BREAK
DAMAGE RESPONSE NETWORKS
Within the living cells, genetic information encoded in DNA sequences is constantly
under threat by various genotoxic stresses. Of the various types of DNA lesions that
arise, double-strand breaks (DSBs) are the most damaging. In mammalian cells, DSB
can result from exogenous agents, such as background radiation and environmental
mutagens, or arise from endogenous sources as metabolically produced free radicals.
In addition, DSB may emanate as DNA intermediates, normal or aberrant, in several
specialized cellular functions, such as DNA replication, V(D)J recombination, meiotic
recombination, class switching, and apoptosis.
The consequences of DSBs are of major cost to the cells. If left unrepaired, even one
DSB can be lethal. Loss of continuity between the DNA fragment and the centromere
From: Cancer Drug Discovery and Development
Apoptosis, Senescence, and Cancer
Edited by: D. A. Gewirtz, S. E. Holt and S. Grant © Humana Press Inc., Totowa, NJ
243
244
Part III / DNA Damage Response, Signaling Pathways, and Tumorigenesis
Table 1
Classification of DNA repair factors according their role in DSB cellular response.
Factor
Participation in
multiprotein
complex
H2AX
Nucleosome
Chromatin structure
Methyl-H3
Lys 79 and
relevant
residues
53BP1
Nucleosome
Chromatin structure
–
MDC1
–
Mre11
Rad50
MRN
MRN
NBS1
MRN
ATM
Homodimer
RAD17
Several complexes
ATRIP
ATR–ATRIP
complex
ATR–ATRIP
complex
DNA-PK
DNA-PK
ATR
DNA-PKcs
Ku 80/70
Chk1
Chk2
BRCA1
BRCA2
P53
Cdc25A
Cdc25C
RAD17
ARTEMIS
XRCC4
LIG4
DSS1
SMC1
–
–
BRCA1–BARD1
and several other
complexes
BASK and several
other complexes
–
–
–
Several complexes
–
XRCC4–LIG4
complex
XRCC4–LIG4
complex
19S proteasome
Cohesin complex
Function
Activator,
adaptor–mediator
Activator,
adaptor–mediator
Nuclease
DNA structural
protein
Activator,
adaptor–mediator
Kinase
Role in DSB
response
Chromatin mark,
transducer
Chromatin mark
Sensor,
transducer
Transducer
Sensor, effector
Effector
Transducer
Sensor (?),
proximal
kinase
Sensor
Chromatin-binding
protein
Adaptor–mediator
Transducer
Kinase
Proximal kinase
Kinase
Sensor
Proximal kinase
Sensor,
transducer
Effector kinase
Effector kinase
Transducer,
effector
Kinase
Kinase
E3 ubiquitin ligase,
scaffold protein,
etc.
Scaffold protein,
etc.
Transcription factor
Phosphatase (?)
Phosphatase
Chromatin binding
Nuclease
Adaptor (?)
Effector
Effector
Effector
Sensor
Effector
Adaptor (?)
Ligase
Effector
Catalytic subunit
Structural protein
Effector
Effector
Transducer
Chapter 13 / Interplay Between H2AX and 53BP1
245
during mitosis can result in deletions of a large proportion of the genetic information
passed to the progeny cells. If not properly repaired, DSBs can cause alterations in
the DNA sequence, chromosomal translocations, genomic instability, and eventually
neoplastic transformation. Thus, it is not surprising that the DSB damage response
is a highly sophisticated set of reactions that channels cellular activities toward three
directions: DNA repair, cell-cycle arrest, and apoptosis.
Two major pathways orchestrate the DSB repair in mammalian cells, one based
on homologous recombination (HR), and the other on non-homologous end joining
(NHEJ). HR promotes accurate repair by copying information from an intact homologous DNA template, preferentially the sister chromatid when available, and predominates in S/G2 phases (1). NHEJ is independent of homology, or utilizes microhomology
to join broken ends, is error-prone, and predominates in G1 (2–4). To provide damaged
cells sufficient time to repair, DNA DSB repair systems initiate signal-transduction
pathways to activate G1/S, intra-S, and G2/M cell-cycle checkpoints. In the presence
of irreparable or excessive damage, checkpoint signaling can also induce apoptotic cell
death pathways (5).
To accomplish its multifaceted role, DSB damage response is organized in networks
that set in motion cellular subroutines. These networks operate via protein-protein interactions that are communicated by post-translational modifications, mainly phosphorylation and dephosphorylation (6).
Conceptually, the DSB damage response networks need to serve three major necessities: the recognition of the damaged site, the efficient amplification of the damage
signal and its transmission to other cell compartments, and the adequate commencing
of cellular subroutines that are going to carry out the end results. Reflecting these
necessities, the proteins that have evolved to participate in these networks fall into
three general classes: sensors, transducers, and effectors. These days, ongoing research
reveals that DSB damage response proteins may participate in more than one class,
demonstrating that the DSB damage response networks are characterized by elevated
sophistication (Table 1).
2. CHROMATIN-BASED EVENTS IN RESPONSE TO DSBs
The lodging of genetic material in the nucleus of eukaryotic cells requires an
extreme compaction of DNA that is achieved at different levels of chromatin folding.
At the nucleosome level, 147 base pairs of DNA are wrapped 1.7 times around the
histone octamer. Further compaction is achieved by the linker histone H1. Nucleosome
arrays are further folded into progressively higher-order structures, with the support
of non-histone structural proteins. Although the structure of the nucleosome is well
characterized, less is known about the molecular nature of more highly folded structures. In this chapter, we focus on two chromatin-based events, H2AX and 53BP1
focus formation, and their role in the biology of DSBs.
After irradiation, DSB lesions can be visualized as distinct formations called foci,
when immunocytochemistry methods are combined with epifluorescent microscopy (7).
Currently it is becoming clear that DSB-dependent foci are dynamic chromatin structures juxtaposed to the lesion, where repair, signal transduction, chromatin, and structural proteins are bound onto DNA. Many repair and signaling factors are known
to translocate to DSB-dependent foci in a time-dependent manner. These factors are
246
Part III / DNA Damage Response, Signaling Pathways, and Tumorigenesis
afterward released from the focus to perform functions in the nucleoplasm or cytoplasm.
Experimental evidence have substantiated the classification of these factors to the three
categories mentioned above, sensors, transducers, and effectors.
Foci facilitate repair and amplification of the checkpoint signal by orchestrating
ordered recruitment, assembly, and activation of further repair and signaling proteins;
focus constitution varies considerably. At a given time, different factors can be detected
in a focus population within the same cell, as they accommodate different repair
complexes depending on the repair system involved, the cell-cycle phase, or the timing
of the recruitment and release process. Although the damage response events that occur
in a cell are not restricted to the sites of lesion, chromatin-based foci events appear to
be crucial in the organization and monitoring of the repair process that takes place at
the damaged sites.
2.1. The Biology of H2AX Foci
2.1.1. H2AX Foci
The “higher-order chromatin domains model” of -phosphorylated chromatin
The first indication that higher-order chromatin structures are involved in the biology
of DSBs was in the case of the phosphorylation of the histone variant H2AX (8).
Upon the introduction of DSBs into genomic DNA, a characteristic SQ motif that is
accommodated in the C-terminus of the histone H2AX becomes phosphorylated in
nucleosome arrays that span a distance of up to megabase-long domains (7).
When immunocytochemistry methods are applied to irradiated cells, and cell
specimens are observed under the fluorescent microscope, H2AX forms large, bright,
and discrete foci at a random distribution throughout the nucleus but not within the
nucleoli area. Focus formation pattern follows fast kinetics; H2AX foci appear as
small and numerous within 1–3 min, become fewer in number but larger and better
detected at 15 min, stay steady in size and number between 15 and 60 min, decrease in
number at 180 min, and eventually almost disappear at 24–48 h. Various normal and
cancer cell lines as well as living organisms respond by the formation of H2AX to
lethal and non-lethal amounts of ionizing radiation (IR) (7).
The demonstration of precise H2AX localization to the sites of DNA DSBs
was achieved by the means of a laser scissors experiment, where DSBs were introduced through a pulsed laser microbeam driven along a predetermined course. Subsequent immunocytochemistry showed that H2AX forms precisely along this track of
DSBs (7).
Remarkably, the phosphorylation of this characteristic SQ motif is conserved over
evolution and is part of the DSB damage response. The histone SQ motif is present
in different species but resides on different histone variants, all members of the H2A
family. In mammals, Xenopus laevis, and Tetrahymena thermophila, the SQ motif
resides on H2AX, in Drosophila melanogaster on H2AvD, and in Saccharomyces
cerevisiae and Schizosaccharomyces pombe on H2A1 and H2A2 (9). In this chapter,
for simplicity and clarity purposes, we are going to use indistinguishably the H2AX
designation, when we refer to general, conserved features of this cellular response,
except where otherwise indicated.
In mammals, the percentage of H2AX with respect to total H2A differs between cell
lines, spanning from 2.5 to 30%, whereas in yeast, the ortholog H2A comprises 95%
Chapter 13 / Interplay Between H2AX and 53BP1
247
of the whole complement. Exploiting this divergence, the Bonner laboratory noted a
very interesting correlation: although a different amount of H2AX is phosphorylated
among different cell lines, the percentage of H2AX versus H2A per DSB is constant.
On the basis of this correlation, the Bonner laboratory proposed a model that predicted
that megabase-long domains in chromatin become -phosphorylated per DSB (8)
(Fig. 1A and B).
The expansion of -phosphorylated chromatin along megabase-long domains model
is based on average values. It predicts a distribution of -phosphorylated nucleosomes along an average domain size rather than a “fixed” length on chromatin.
In mitotic Muntiacus muntjak cells from cultures that were previously exposed to
sublethal amounts of IR, H2AX foci form band-like structures on chromosome
arms (7). From a different end, immunoprecipitation experiments in yeast with a
H2AX antibody revealed that -modified chromatin extends in the range of several
hundred kilobases (10). In budding yeast, it has been showed that a single DSB induces
the -formation of an approximately 100-kb domain around the lesion (11). It is very
likely that these domains reveal higher-order chromatin structures that may be characteristic of the particular species. Yet, little is known about the organization of these
chromatin domains.
A
B
ATM/ATR/DNA-PK
C
E
D
MDC1
MDC1 FHA
BRCT
MRN
?
MRN
?
Fig. 1. -Phosphorylation alters the affinity properties of megabase-long chromatin domains. When
double-stranded DNA damage occurs by irradiation (A) or other factors, multiple members of the
phosphatidylinositol-3 (PI3) kinases family, namely ataxia telangiectasia mutated (ATM), ATM and
Rad3 related (ATR), and DNA-depended protein kinase (DNA-PK), redundantly phosphorylate Ser
139 residue of the histone H2AX (H2AX) (B). -phosphorylation expands along megabase-long
chromatin domains, reveals the existence of higher-order chromatin structures that are involved in the
biology of double-strand break (DSB) repair, alters the affinity properties of the affected chromatin,
and depicts a biological amplification mechanism as DSB sites are surrounded by thousands of
-modified nucleosomes (C). MDC1 translocates to the lesions sites and binds to H2AX through
its BRCA1 carboxy-terminal (BRCT) domains (D). The MRN complex (Mre11, Rad50, and NBS1)
translocates to the lesions sites and binds to damaged DNA through its Mre11 subunit. Forkheadassociated (FHA) domains of NBS1 may facilitate a direct or indirect association of MRN with
H2AX. Subsequently, NBS1 recruits and activates ATM at the damage site (E).
248
Part III / DNA Damage Response, Signaling Pathways, and Tumorigenesis
The expansion of -phosphorylated chromatin along megabase-long domains depicts
a biological amplification mechanism where the DSB site is surrounded by thousands
of -modified nucleosomes (Fig. 1). A practical implication of this amplification is
the notion that even one DSB can be visualized by immunocytochemistry methods.
Indeed, during V(D)J recombination, recombination-activating gene (RAG)-mediated
cleavage generates DBSs between immunoglobulins and T-cell receptor loci (12,13,14).
In developing thymocytes, H2AX forms nuclear foci that colocalize with the T-cell
receptor locus, as determined by immunofluorescence in situ hybridization (14).
This result also demonstrates that immunocytochemistry with H2AX antibodies is
a powerful tool as it can detect the presence of only one DSB per nucleus.
2.1.2. Kinases Responsible for -Phosphorylation of H2AX
Phosphorylation of H2AX at Ser 139 is governed by multiple kinases that are
members of the phosphatidylinositol-3 (PI3) family, namely ataxia telangiectasia
mutated (ATM, ATM and Rad3 related (ATR), and DNA-depended protein kinase
(DNA-PK). In S. cerevisiae, Tel1 and Mec1, the yeast homologs to ATM and ATR,
respectively, are also involved in -phosphorylation of H2A, which is the yeast
homolog to H2AX (9,15). When yeast strains that bear deletions in tel1 or/and mec1
genes are subjected to methyl methane-sulfonate (MMS) treatment that involves generation of putative DSBs, -phosphorylation of H2A is impaired (16). A low-level signal
is present in the mec1 or tel1 null single mutants, but no signal is detectable in
tel1/mec1 double null (16). The above results indicate that both kinases are involved
in H2A phosphorylation and have overlapping roles. In addition, immunoprecipitated
Mec1 from yeast cell extracts can phosphorylate a C-terminal yeast H2A peptide in
vitro, indicating that Ser 129 of the SQ motif of the yeast H2A is a direct target
for Mec1 (10).
In human cells, ATM seems to be the major kinase that controls -phosphorylation
(17–19). ATM knockout cells exhibit impaired -H2AX focus formation that can
be further eliminated by low-concentration treatment of wortmannin, indicating a
redundant role of DNA-PK and/or ATR (20). Low doses of IR activate ATM
to -phosphorylate H2AX, whereas at higher doses, other kinases contribute or
substitute. Fluorescent microscopy of ATM colocalization at the sites of DSBs has
been problematic because of the abundance of the former molecules throughout the
nucleus. Retention of the ATM molecules at DSBs and colocalization with H2AX are
shown only after in situ extraction of the unbound ATM molecules before immunocytochemistry (21). In accordance with the in vivo experiments, immunoprecipitated
ATM phosphorylates the SQ motif of H2AX on Ser 139 in vitro (17).
Under hypoxic conditions, H2AX is -phosphorylated in an ATR-dependent
manner (22). In S phase of the cell cycle, ATR is the kinase to take over
-phosphorylation, in response to replication arrest and the consequent generation of
DSBs, or upon formation of topoisomerase I cleavage complexes after collision of DNA
replication forks (23,24). Nevertheless, ATM and ATR exhibit extended redundancy
as these kinases are known to have several targets in common, including H2AX (25).
The role of DNA-PK in -phosphorylation of H2AX is not yet well understood.
Although crude extracts of DNA-PK phosphorylate the SQ motif of H2AX on Ser
139 in vitro, several rodent cell lines that are deficient either in the DNA-PK catalytic
subunit or in the Ku antigen exhibit no detectable defects in -phosphorylation of
Chapter 13 / Interplay Between H2AX and 53BP1
249
H2AX (8,26). However, it has been reported that DNA-PK plays a redundant role in
several cases as shown in astrocytoma M059J cell line (7,27) and upon formation of
topoisomerase I cleavage complexes at replication forks in S phase (24).
By fluorescence microscopy, DNA-PK immunocytochemistry reveals a diffuse
pattern throughout the nucleus in both non-irradiated and irradiated cells. However,
upon ionizing irradiation, the DNA-PK catalytic subunit becomes autophosphorylated
on threonine 2609 and colocalizes with H2AX in distinct foci (28). Autophosphorylation of the DNA-dependent protein kinase catalytic subunit is required for rejoining
of DNA DSBs, but the involvement of H2AX in this mechanism is not clear yet.
H2AX was found to be essentially immobile in chromatin (29). N-terminal fusion
constructs of H2AX with green fluorescent protein (GFP) were used to study their
diffusional mobility in transient and stable cell transfections. In the absence or
presence of DSBs, only a small fraction of GFP-H2AX was redistributed after photobleaching. This fact suggests that a phosphorylation–dephosphorylation cycle takes
place during the induction and disappearance of foci rather than that a diffusionexchange mechanism (29).
2.1.3. The H2AX DSB-Repair Pathway
Colocalization and interactions of repair and signaling factors with H2AX foci
Many components of the DNA damage response, including ATM, 53BP1, MDC1,
MRN complex, BRCA1, and SMC1, form IR-induced foci (IRIF) that colocalize with
H2AX foci (9,30–32). These factors participate in HR and NHEJ, indicating that
H2AX plays a role in both repair systems in mammals. It has been demonstrated that
there is a time-dependent sequential assembly of repair factors and signal mediators
on H2AX foci (27). When cells are treated with wortmannin, a known inhibitor of
the PI3 kinase family, H2AX focus formation is abolished and 53BP1, MNR, and
BRCA1 foci are severely impaired. Along the same line, in H2AX−/− cells, initial
migration of 53BP1 and MNR to IRIF is not totally abrogated, but further accumulation
is diminished (33). This phenotype rises the question as to whether H2AX is a crucial
component of double-strand repair and what role it plays in this response.
The Nussenzweig laboratory has proposed the “two-stage recruitment model,”
according to which H2AX does not constitute the primary signal that is required
for the redistribution of repair complexes to damaged chromatin, but functions as a
platform to concentrate repair factors to the vicinity of DNA lesions and to promote
interactions between multicomponent complexes (33). The accumulation of repair and
signaling factors in proximity to a DSB would facilitate an amplification step of signal
transduction and checkpoint pathways, particularly in the case where low numbers
of foci per nucleus are present in cells. H2AX also modulates the accumulation of
repair/signaling proteins in chromatin regions distal to a DSB, following their initial,
H2AX-independent migration to DSBs. The retention and subsequent increase in the
local concentration of factors may be mediated through weak interactions between the
SQ motif in the H2AX tail, thousands of which are modified by phosphorylation, and
specific domains of repair/signaling proteins.
This model is also consistent with the finding that H2AX−/− cells exhibit reduced
ability to arrest the cell cycle at low doses of IR (33). In the case when only a few DSBs
are generated in the nucleus, the DNA repair factors that are modified to transduce
the signal are limited, and signal amplification at IRIF becomes essential. On the
250
Part III / DNA Damage Response, Signaling Pathways, and Tumorigenesis
contrary, the haploinsufficiency of H2AX+/− mice (13,34), which is based on dosage
dependence of the H2AX gene, could be explained with the hypothesis that chromatin
of H2AX+/− cells, which comprises sparsely H2AX-containing nucleosomes, would
not mediate efficient concentration of soluble DNA repair factors on IRIF and further
amplification of the signal. Notably, in experimental models where Ser 136 and Ser
139 were substituted with glutamic acid, IRIF fail to form (34). Moreover, these cells
exhibit enhanced sensitivity to IR comparable with the wild type, indicating that the
actual phosphoserine 139 is essential for biological activity and cannot be substituted.
2.1.3.1. H2AX Interactions with MDC1. The mediator of DNA damage checkpoint
protein 1 (MDC1) is a novel nuclear protein that contains a forkhead-associated (FHA)
domain and two BRCA1 carboxy-terminal (BRCT) domains and possesses in vitro
DNA binding activity (35). MDC1 also presents 20 potential ATM/ATR consensus
target phosphorylation motifs (Ser/Thr-Gln), located throughout its N-terminal half,
and another 19 consecutive imperfect repeats in its central region (36,37).
In irradiated cultured human cells, MDC1 translocates to the lesion sites rapidly
to form foci. MDC1 foci colocalize extensively with H2AX foci and exhibit similar
kinetics (37–39). It has been shown that MDC1 localization to the DSB sites is
abolished in H2AX−/− mouse embryonic fibroblasts (36). In support of this observation,
MDC1 foci are also abolished in human cells where H2AX expression had been downregulated by small interfering RNA (siRNA). These experimental results demonstrate
that MDC1 focus formation is strictly dependent on H2AX and place these factors in
the same pathway (40,41).
BRCT domains are considered to account for the focus formation of MDC1, as
in experiments where these domains were deleted, MDC1 foci were compromised
(42). As it has been shown that BRCT domains recognize phosphopeptides (43), it is
reasonable to speculate that these domains recognize the H2AX tail and drive toward
the MCD1 focus formation (Fig. 1). In support to this speculation, in vitro experiments
have shown that a H2AX C-terminus peptide phosphorylated on Ser 139 interacts
with MDC1 in human cell extracts, whereas the equivalent non-phosphorylated peptide
does not (44).
Down-regulation of MDC1 expression levels by siRNA renders cells hypersensitive to IR (36). At the cellular level, MDC1 down-regulated cells exhibit defects in
IR-induced G2/M and intra-S phase checkpoints and show reduced levels of IR-induced
apoptosis (45). At the molecular level, the MRN complex and BRCA1 fail to efficiently
accumulate in IRIF (40,46). 53BP1 accumulation at sites of DNA damage was found
to be MDC1 independent by three different research groups (36,38,47) and partially
dependent by another (39).
2.1.3.2. H2AX Interactions with MRN Complex. Colocalization of H2AX foci
with the three-protein MRN complex is evident in IRIF linking H2AX with both DNArepair and checkpoint cell-cycle responses (24,27,48). Mre11, Rad50, and NBS1 play
a crucial role in the biology of DSB repair and operate together in a form of a stable
complex (49).
Mre11 has a C-terminal DNA-binding domain and possesses a 3 –5 exonuclease
activity, a single-strand endonuclease activity, and a limited DNA unwinding activity
(50). Mre11 is characterized by high affinity for aberrant DNA structures.
Chapter 13 / Interplay Between H2AX and 53BP1
251
Rad50 is an ATPase with a characterized structural role; Rad50 binds DNA through
a globular domain, whereas the coiled-coil regions of Rad50 form an extended
intramolecular flexible arm. It has been shown by force scanning microscopy that
Mre11 and Rad50 heterodimers seriate along broken DNA fragments (51). The
intramolecular flexible arm of Rad50 on one DNA fragment “weaves” with the reciprocal Rad50 arm of a distant DNA fragment to hold the two molecules together. It is
postulated that Rad50 gives the MRN complex the ability to tether sister chromatids
together in vivo (52).
The NBS1 role as a mediator in the DSB damage response is well established (53).
NBS1 protein has an N-terminal FHA domain and a BRCT domain (54). In H2AX
knockout cells, NBS1 foci (and presumably MRN complex foci) are impaired but
not abolished, indicating that H2AX does not comprise the primary signal for the
recruitment of MRN complex at the lesion sites (33). These results indicate that the
MRN complex has properties of a sensor protein through the DNA-binding domain of
Mre11 (49).
H2AX-modified chromatin plays a crucial role in the accumulation of NBS1 at
the lesion sites (40). In vivo disruption of FHA domains abrogates its interaction with
-modified chromatin, indicating a direct or indirect association of NBS1 with H2AX
(54) (Fig. 1). Moreover, NBS1 interaction with H2AX is independent of hMre11 or
BRCA1 (55). Interestingly, MDC1-depleted cells fail to accumulate the MRN complex
in the vicinity of the DSB damaged sites, indicating that this interaction is mediated
by MDC1 (see Section 2.1.3.3.).
Several consensus ATM phosphorylation SQ motifs are located within the central
region of NBS1 (56). In particular, Ser 278, Ser 343, and Ser 397 are phosphorylated
by ATM in response to irradiation. In cells that express mutations at these sites to
prevent phosphorylation, no difference was observed in the ability of NBS1 to form
foci. Although the role of these phosphorylation events is not clear yet, it has been
postulated that they function downstream to mediate the signal to other ATM substrates
(55,57–59).
It has been demonstrated that migration of NBS1 molecules occurs independently
of their phosphorylation status. A strong indication that NBS1 phosphorylation takes
place at the foci, where ATM also migrates in response to DSB, is derived from
studies in cells where NBS1 was fused to histone H2B. It was shown that NBS1-H2B
molecules were phosphorylated after irradiation only at the lesion sites and not in the
residual, undamaged chromatin (40).
NBS1 has an additional function, as activator of ATM. ATM autophosphorylation
in response to IR is profoundly impaired in human Mre11ATLD1-expressing cells that
contain a C-terminally truncated Mre11 domain (58,60,61).
2.1.3.3. Tripartite Interactions Between H2AX, MDC1, and MRN Complex.
At present, the interaction between MDC1 and MRN complex is being actively
investigated. Down-regulation of MDC1 by siRNA reveals a similar phenotype as in
the H2AX knockout and knockdown cells, characterized by impaired, but not abolished
NBS1 focus formation (36,38,62). Taking into account that MDC1 focus formation is
strictly dependent on H2AX (see Section 2.1.3.3.), it is reasonable to speculate that
the contribution of H2AX to the accumulation of MRN complex at the lesion sites is
mediated by MDC1 (Fig. 2). Consistent with its role as a mediator, FHA domains of
252
Part III / DNA Damage Response, Signaling Pathways, and Tumorigenesis
A
MDC1
MDC1
MDC1
MDC1
MDC1
MDC1-
MDC1
MDC1
B
ATM
activated ATM
MRN
MRN
MRN
?
MDC1
MDC1
MDC1
MDC1
MDC1
MDC1
MDC1
MDC1
Fig. 2. MDC1 and MRN create a reinforcement loop to enhance ATM activity on -modified
chromatin. MDC1 relocates to H2AX foci and binds to phosphorylated Ser 139 through its BRCA1
carboxy-terminal (BRCT) domains (A), where it creates a scaffold to facilitate anchorage of MRN
complex (B). In turn, MRN complex enhances ATM activity to facilitate phosphorylation of repair
factors, including MRN and MDC1. Multiple phosphorylation of MDC1 may promote multimerization of this protein through its forkhead-associated (FHA) domains and facilitates anchorage of
other repair factors that are subsequently recruited to the lesion sites.
MDC1 may be involved in the multimerization of this protein on the DSB-modified
chromatin. It is possible that MDC1 multimerization could function to create a scaffold
on -modified chromatin, where the MRN complex and other checkpoint and DNA
repair proteins could anchor and interact with other factors (37,38,40) (Fig. 2).
In cells expressing FHA-disrupted NBS1, NBS1 accumulation was abrogated, but
microfocus-like structures were formed along the DSB path that was generated by
microlaser (40,47,54). These structures are MDC1 independent, demonstrating that
there is a small region at the lesion sites where NBS1 associates in a FHA-independent
way. Whether these results indicate that NBS1 binds directly to naked DNA or interacts
with -modified chromatin through alternative means remains to be experimentally
approached (Fig. 1).
MDC1 becomes phosphorylated in response to IR in an ATM-dependent manner.
This phosphorylation is partially affected in H2AX−/− MEFs, and NBS cells,
suggesting that both NBS1 and H2AX participate in a feedback loop reaction
involving ATM activation (41,55). These results are consistent with a model according
to which MDC1 relocates to H2AX foci, where it attracts the MRN complex. In
turn, NBS1 activates ATM that is also relocated there, which subsequently phosphorylates MDC1 (Fig. 2). In support to this model, -phosphorylation is also affected in
MDC1-depleted cells (35). However, the possibility that MCD1 phosphorylation could
Chapter 13 / Interplay Between H2AX and 53BP1
253
occur in parallel at the nucleoplasm by other mechanisms cannot be excluded from
these experiments.
In addition to its role as mediator, MDC1 has a function as an activator to ATM as
well (see Section 3).
2.2. The Biology of 53BP1 Foci
2.2.1. 53BP1 Foci
53BP1 was originally identified during a yeast two-hybrid screen for p53-interacting
proteins (63). The sequence similarity of 53BP1 with the yeast DNA damage checkpoint proteins Rad9 and Crb2/Rhp9 and the localization of 53BP1 to sites of IRIF
suggested that 53BP1 would function in the DNA DSB checkpoint pathway. The first
experimental evidence showing that 53BP1 is involved in the cellular response to
DNA damage was its localization to sites of DSBs in cells exposed to IR (64). Upon
DNA damage, 53BP1 relocalizes to discrete nuclear foci that represent sites of DNA
lesions and become hyperphosphorylated (65,66). The notion that 53BP1 foci mark
sites of DNA DSBs has been documented by colocalization experiments of 53BP1
with H2AX and other repair factors known to form foci (62,64–68).
In G1, 53BP1 exists in a diffuse nuclear pattern as well as in large nuclear “dots.” In
S phase, 53BP1 can be found in a discrete, punctuate pattern. The nuclear distribution
pattern of 53BP1 in G2 cells appeared in two forms, one similar to S phase but with
fewer foci and one that exhibited few, if any, large dots. 53BP1 foci form within
minutes of irradiation, and at doses of IR as low as 0.5 Gy. 53BP1 foci colocalize with
H2AX and exhibit kinetics parallel to H2AX focus formation (64,69). The number
of 53BP1 foci increases linearly over time, reaching a maximum at about 15–30 min,
and then steadily decreases to baseline levels within the next 16 h. The number of
53BP1 foci, about 20 per cell per Gy of IR, closely parallels the number of DSBs. In
addition, the kinetics of resolution of 53BP1 foci is very similar to the kinetics of DNA
DSB repair following IR, indicating that there is a relationship between completion of
repair and disappearance of 53BP1 foci (64,70).
Several domains have been characterized in 53BP1: a 53BP1 focus formation region
(FFR), two tandem tudor folds, two tandem BRCT domains, and an 8-kDa light chain
(LC8)-binding domain (71–76). A nuclear localization signal is also present. A 53BP1
FFR has been mapped at residues 1052–1639 (69,77). It was previously shown that
H2AX coimmunoprecipitates with 53BP1 from irradiated cells (69). Interestingly,
in a pull-down assay, where six different 53BP1 GST fragments spanning the entire
53BP1 protein were incubated with immobilized C-terminus H2AX peptides, only the
fragments 956–1354 showed strong interaction with the -phosphorylated one (72,78).
In contrast, no binding was detected to the non-phosphorylated peptide bearing identical
sequence.
At the C-terminus of the FFR, there are two tandem tudor folds that consist of a 50
amino acid long stretch (72,76). Although not well established, Tudor domains seem
to play a role in protein–protein interactions through methylated residues. In support to
this notion, tudor folds have structural similarities to chromo domains, that are known
to bind to histone tails that contain methylated lysines (see Section 2.2.2.).
The FFR also includes a region required for 53BP1 kinetochore localization
in mitotic cells (amino acids 1220–1601) (76). In a different set of experiments
254
Part III / DNA Damage Response, Signaling Pathways, and Tumorigenesis
where various GFP-tagged 53BP1 truncations were expressed, the kinetochore-binding
domain has been mapped to a 380 residue portion of the protein that excludes the
nuclear localization signal and the BRCT motifs (79). It was shown that this region
is responsible for loading 53BP1 to kinetochores in prophase and release by midanaphase (79). Recently, it was shown that the kinetochore localization region is also
essential for 53BP1 focus formation in response to DNA damage, suggesting that both
events might be regulated in a similar fashion.
An LC8-binding domain of 53BP1 has been mapped to a short peptide segment
immediately next to N-terminal to the kinetochore localization region (75). Unlike
other known LC8-binding proteins, 53BP1 contains two distinct LC8-binding motifs
that are arranged in tandem. The LC8-binding domain is completely separated from
the p53-binding domain in 53BP1. As the LC8 of dynein binds to 53BP1, it has been
proposed that 53BP1 can potentially act as an adaptor to assemble p53 to the dynein
complex (75).
BRCT domains are a common protein–protein interaction motif that are present
in proteins involved in the DNA damage response. 53BP1 has two tandem BRCT
domains (amino acids 1714–1850 and 1865–1972, respectively) (64,66,73). A role for
53BP1 in a DNA damage response pathway was first proposed based on the similarity
of its BRCT domains to the BRCT domains present in the S. cerevisiae Rad9 and
S. pombe Crb2/Rhp9 proteins (64,71,72). Rad9 in S. cerevisiae is required for cellcycle arrest in response to DNA damage and becomes phosphorylated by Mec1. After
phosphorylation, Rad9 interacts with Rad53, which is the homolog of the Chk2 kinase
in humans, and this interaction is required for the activation of Rad53.
A comparison of the structure of the BRCT region of 53BP1 with the BRCT tandem
repeats reveals that the interdomain interface and linker regions are remarkably well
conserved. The crystal structure of the 53BP1 BRCT tandem repeats in complex with
the DNA-binding domain of p53 shows that the two tandem BRCT repeats pack
extensively through an interface that also involves the inter-repeat linker (72,74). The
first BRCT repeat and the linkers together bind p53 on the region that overlaps with
the DNA-binding surface of p53 and involves p53 residues that are mutated in cancer
and are important for DNA binding.
It has been suggested that the interaction of p53 with 53BP1 reflects the putative
adaptor function of 53BP1; 53BP1 may recruit p53 to sites of DSBs, thereby
facilitating its ATM-dependent phosphorylation (72). However, this is not the only
possible scenario. Recent experimental results suggest that BRCT domains may have
phosphopeptide-binding activity, depending on the structure in the vicinity (43). Given
the absence of p53 in yeast and the evolutionary conservation of the BRCT domains of
53BP1 in all eukaryotes, the core function of the 53BP1 BRCT domains could relate
to some function of 53BP1 that are independent of p53 (72,78,80,81). Because the
BRCT domains are not required for localization of 53BP1 to sites of DNA DSBs, it
is likely that they play a role in ATM activation, directly or indirectly (41,71,82). The
identification of a physiological relevant phospholigand of the 53BP1 BRCT domains
will shed light to this question.
Fifteen potential phosphorylation sites (15AQ) have been mapped on 53BP1
N-terminal region, some of which are known to be targeted during the DNA damage
response (66). Indeed, 53BP1 becomes hyperphosphorylated on its N-terminus in
Chapter 13 / Interplay Between H2AX and 53BP1
255
an ATM-depended manner in response to IR and mediates DNA damage-signaling
pathways in mammalian cells (41,66,78,81,82).
2.2.2. Recruitment of 53BP1 to Sites of DNA DSBs
As mentioned in Sections 2.1.3. and 2.1.3.1., in cells that lack histone H2AX,
or where the phosphorylation of H2AX is abrogated, 53BP1 is initially recruited
to the damaged chromatin, but cannot be retained (33). A number of laboratories
have addressed the question regarding the nature of the factor that recruits 53BP1 to
DSBs sites (47,72). The initial recruitment of 53BP1 to the site of a DSB does not
seem to require ATM, NBS1, or DNA-PK, since in cells deficient in these proteins,
53BP1 localizes to sites of DSBs with normal kinetics (33). Recently, a role of
histone methylation in the DSB damage response has been discovered. The Halazonetis
laboratory showed that in human cells, 53BP1 binds to methylated H3 (72). In vitro
experiments using residues that form the walls of the pocket between the tudor folds
showed that the 53BP1 tudor domain binds histone H3 methylated on Lys 79 (72).
By deletion analysis, these residues were also shown to be required for recruitment
of 53BP1 to DSBs in vivo. Competition experiments with H3 peptides showed that
the H3 peptide with dimethylated Lys 79 competed with the native histone H3 (72).
The corresponding non-methylated peptide did not compete, whereas the peptides with
mono-methylated and dimethylated Lys 27, or Arg 26 competed at lower efficiency
(72). The enzyme that methylates Lys 79 in human cells is DOT1, an evolutionary
conserved methyltransferase. Suppression of DOT1 by siRNA resulted in suppression
of H3 methylation on Lys 79 and minimized the 53BP1 recruitment to the lesion sites.
One possible explanation for the reduced affinity of 53BP1 to the damaged chromatin
when abolished either the relevant methylated sites or the -phosphorylation could
be that the tudor domain and the phosphopeptide-binding region act in a cooperative
way. As mentioned earlier in Section 2.2.1., the tudor domain and the phosphopeptidebinding region reside in proximity within the FFR of 53BP1, supporting further this
hypothesis (Fig. 3).
Surprisingly, the levels of H3 Lys 79 methylation were unaltered upon DSB
damage (72). To resolve this discrepancy, the following model was proposed by the
Halazonetis laboratory. The introduction of DSBs into chromatin results in disruption of
nucleosome stacking, which leads to revelation of H3 Lys 79 and possibly other methylated residues of the histones in the core particle, resulting in exposure of binding sites
for 53BP1 and its recruitment to the DSB sites. According to this proposed mechanism,
53BP1 can “sense” DSB lesions through changes in higher-order chromatin structure
and participates as a sensor in the DSB damage response (72).
This model, based on the results from human cells, has been supported by parallel
work in S. cerevisiae and S. pombe. In S. cerevisiae, Rad9, a DNA damage checkpoint
protein that has a sequence similarity to human 53BP1, interacted with native H3 (83).
Deletion of DOT1 in yeast resulted in radiation sensitivity and a DSB checkpoint
defect. In the fission yeast S. pombe, it was demonstrated that Set9, a previously
uncharacterized SET domain protein, is responsible for H4 Lys 20 methylation (84).
Interestingly, this methylation does not have any apparent role in the regulation of gene
expression or heterochromatin function. However, loss of Set9 activity or mutation of
H4 Lys 20 markedly impairs cell survival after genotoxic challenge and compromises
the ability of cells to maintain checkpoint-mediated cell-cycle arrest. Furthermore,
256
Part III / DNA Damage Response, Signaling Pathways, and Tumorigenesis
A
B
ATM
C
D
p53BP1
activated ATM
p53BP1
Fig. 3. Recruitment of 53BP1 to damaged sites by chromatin -phosphorylation and methylation.
Collectively with -phosphorylation, the introduction of double-strand breaks (DSBs) into chromatin
results in disruption of nucleosome stacking (A and B), which leads to revelation of methylated
H3 lysine 79 and possibly other methylated residues of the histones in the core particle (B). 53BP1
relocates to the damaged chromatin where it binds through its focus formation region (FFR). FFR
includes a phosphopeptide-binding site and a tudor domain. In the absence of H2AX or the relevant
methylated residues, 53BP1 is recruited to the DSBs sites but not retained, indicating that FFR
facilitates cooperative binding (C). 53BP1 creates a scaffold on damaged chromatin to recruit and
activate ATM. Activated ATM phosphorylates repair factors including p53BP1 (D).
genetic experiments have demonstrated that Set9 is required for Crb2 localization to
sites of DNA damage (84). Although H4 Lys 20 is a different methylation site from
H3 Lys 79, it seems that these sites play the same role with respect to the 53BP1/Crb2
recruitment mechanism. This could be explained if the affinity of these factors is
not restricted to one methylated site, or multiple histone “marks” may contribute to
53BP1/Crb2 recruitment.
It needs to be mentioned also that BLM helicase has been reported to recruit 53BP1
to DSB lesions independently of its helicase activity (67). Interestingly, a deficiency
of the BLM helicase has been reported that markedly increases cancer incidence in
humans (85).
3. INTERPLAY BETWEEN H2AX AND 53BP1
To recapitulate, the introduction of DSBs into DNA results in long stretches of
chromatin that are marked by two histone modifications: histone methylation (mainly
the H3 Lys 79 in mammals and H3 Lys 20 in S. pombe) and the -phosphorylation
of H2AX. In turn, this DSB-modified chromatin initiates two branches of the DSB
damage response: one by the 53BP1 recruitment by methylated H3 Lys 79 (or
relevant residues) and the other by the recruitment of MDC1/MRN complex by
H2AX (31,33).
The H2AX and the 53BP1 branches do, however, clearly interact with each other.
The lack of retention of 53BP1 in H2AX−/− cells may indicate a direct mechanism
where the FFR of 53BP1 facilitates cooperative binding of Ser 139 phospho groups and
H3 Lys 79 methyl groups. Nevertheless, H2AX seems to play a role in accumulation
of 53BP1 by indirect mechanisms as well.
Chapter 13 / Interplay Between H2AX and 53BP1
257
Notably, phosphorylation of 53BP1 is reduced by 40% in H2AX−/− cells, 1 h after
low-dose irradiation (86). Although it is not clear whether 53BP1 is phosphorylated only
when bound to damaged DNA, these results suggest that H2AX creates a reinforcement
loop to 53BP1 phosphorylation. Notably, in AT cells, 53BP1 relocalizes at DSBs,
indicating that ATM-dependent phosphorylation of 53BP1 is not required for 53BP1
focus formation (69). What role this phosphorylation plays is however not clear yet.
Several lines of evidence have shown that 53BP1 is required for a subset of
ATM-dependent phosphorylation events (71). An antibody that specifically recognizes
the phosho-Ser/Thr-Gln epitope generated by ATM/ATR phosphorylation showed
reduced immunofluorescence reactivity in cells treated with 53BP1 siRNA (41,82).
However, the magnitude of the defects observed in this system was smaller than that
observed in AT cells, indicating the existence of additional 53BP1 network branches
that lead to ATM-dependent phosphorylation. In addition, specific antibodies against
SMC1 phospho-Ser 966 and phospho-Ser 957 and Chk2 phospho-Ser 33 and phosphoSer 35 showed suppressed phosphorylation in 53BP1 knockdown cells (82). The
above mentioned phosphorylation events have been taken as an indication that 53BP1
functions downstream of ATM activation, as a mediator to several ATM substrates.
In support of this conclusion, it has been shown that 53BP1 is required for efficient
accumulation of p53 to DSB sites and subsequent p53 activation (81).
According to this line of evidence, 53BP1 functions by creating a scaffold on the
revealed methylated H3 Lys 79 chromatin that is further stabilized by interactions with
H2AX to recruit ATM substrates. Recruitment of ATM to the lesion sites would
facilitate the phosphorylation of these substrates. This model is consistent with our
current understanding of Rad9 function in budding yeast; Rad9 is phosphorylated by
Mec1 and binds to Rad53 to recruit it to the lesion sites (72).
The same experimental results could, however, be interpreted in a reverse way.
53BP1 may function upstream of ATM to activate it, in response to DSBs. In fact,
a factor that recruits and possibly activates ATM to the DSBs could be 53BP1, as
53BP1 physically interacts with ATM in irradiated, but not in non-irradiated cells,
as shown by coprecipitation experiments (71). Consistent with this hypothesis is the
observation that 53BP1 forms foci at ATM−/− , indicating that it functions upstream
of ATM activation (71). According to this model, 53BP1 accumulation to the lesion
sites would facilitate the recruitment of ATM, where the ATM substrates would be
recruited by H2AX.
The Halazonetis laboratory has proposed a solution that combines features of two
previous opposing models. According to this group, 53BP1 acts both as an activator
of ATM and as an adaptor and/or a mediator. 53BP1 functions creating a feedback
loop upstream of ATM by activating this kinase, and also downstream of ATM by
facilitating the ability of ATM to phosphorylate its substrates, such as Chk1 and Chk2
(72). A dual function of 53BP1 as an activator and adaptor is certainly possible, given
that ATM phosphorylates 53BP1 in response to DNA damage at Ser 25 and most
likely at other sites as well. Whether these phosphorylation events are critical for
phosphorylation of ATM substrates still remains to be demonstrated (87).
To develop experimental support for the “activator” function of 53BP1, the
Halazonetis laboratory addressed the question whether 53BP1 contributes to the
activation of ATM. It has been demonstrated by the Kastan laboratory that ATM can
be activated at a distance from chromatin, in response to the DSB (88). This research
258
Part III / DNA Damage Response, Signaling Pathways, and Tumorigenesis
group has shown that ATM is located at the nucleoplasm and exists in a dimer/oligomer
form in an inactive form. Upon chromatin aberrations, ATM responds by autophosphorylation on Ser 1981. This phosphorylation disrupts the oligomer form and renders
ATM active by exposing its catalytic site (88). Remarkably, ATM is very sensitive
in “sensing” chromatin irregularities, even when they are not derived from DSBs,
indicating a wide-ranging role of this kinase in response to chromatin abberations.
Whether ATM directly senses these DSB-induced chromatin alterations, or requires
DNA “sensors” to transmit the damage signal is not clear.
To test the hypothesis that 53BP1 plays a role in ATM activation, the Halazonetis
group addressed the question whether 53BP1 contributes to the phosphorylation of
ATM at Ser 1981, by suppressing 53BP1 (72). In contrast to what they were expected,
their first observation indicated that there is no effect on ATM Ser 1981 phosphorylation
in 53BP1 knockdown cells.
Analyzing further into their experimental results, they took into consideration that
ATM is activated also by NBSI, as was shown by different research groups (41,56,57,59).
The MCD1/MRN complex is recruited to the lesion sites, and this response is mediated
by H2AX, independently to 53BP1 (59). This could provide an explanation to the
discrepancy between their woring hypothsis and their experimental results, as the
MCD1/MRN contribution could overpower the effect on ATM Ser 1981 phosphorylation.
Exploring the possibility that there are conditions where the two pathways do not
overlap, the recruitment of both factors to IRIF was monitored at high and low doses
of irradiation (41). The results show that exposure of cells to low dose of IR exhibited
recruitment of 53BP1 but not that of MCD1/MRN complex. This piece of evidence
emphasizes the notion that 53BP1 plays an essential role at low doses of irradiation (41).
Further pursuing the same question, these investigators investigated the possibility
that there is crosstalk between the 53BP1 and the MDC1/MRN pathway. At low-dose
irradiation, suppression of MDC1 had no effect on 53BP1 recruitment, but surprisingly, suppression of 53BP1 compels recruitment of MCD1/MRN complex to IRIF
(41). Similar results were observed with fibroblasts derived from an individual with
XPC (xeroderma pigmentosum group C), in which suppression of 53BP1 resulted in
increased localization of MDC1 and NBS1 (41). The increased localization of NBS1
was accompanied by increased phosphorylation of Ser 343, suggesting that in the
absence of 53BP1, a greater pool of NBS1 molecules is recruited (41).
Finally, and in accordance with their previous experiments, the Halazonetis group
showed that suppression of 53BP1 in NBS1-deficient cells revealed a defect in ATM
phosphorylation at low doses of irradiation (41). In addition, downstream targets of
ATM, specifically Chk2 Thr 68 and SMC Ser 957, were also found to exhibit impaired
phosphorylation. This experimental evidence supports a model according to which
53BP1 has a dual role as an activator and as an adaptor and/or a mediator in the DSB
damage response (Fig. 3).
A different line of experiments from the Lukas laboratory indicates another
relationship between 53BP1 and MDC1 at later time points post-irradiation. When
the departure of 53BP1 from H2AX foci was monitored by real-time microscopy,
it was evident that this departure is accelerated in MDC1-depleted cells (47). In
these experiments, GFP-tagged 53BP1 was observed to leave the H2AX foci after
7–12 h post-irradiation. Surprisingly, in MDC1 knockdown cells, the departure time
was reduced to less than 6 h post-irradiation while H2AX was still present at the sites
Chapter 13 / Interplay Between H2AX and 53BP1
259
of the lesion (47). It has been suggested by these investigators that phosphorylation
of MDC1 at foci renders it able to interact directly with 53BP1, or alternatively, with
remodeling factors that stabilize chromatin conformation changes that facilitate DSB
repair (47).
4. OPEN QUESTIONS AND PERSPECTIVES
DSB-dependent foci are dynamic chromatin structures that facilitate repair and
amplification of checkpoint signals by orchestrating a time-dependent ordered
recruitment, assembly, and activation of sensors, transducers, and effectors during the
DSB damage response. Focus constitution varies considerably at a given time, as
different factors can be detected on a focus population within a particular cell, as they
accommodate different repair complexes depending on the repair system involved,
the cell-cycle phase, or the severity of the damage. Although the damage response
events that occur in a cell are not restricted to the sites of lesion, but take place at the
nucleoplasm and cytoplasm as well, focus events appear to be crucial in monitoring
the repair process that take place at the damaged sites.
During this dynamic series of events, a very reliable point of reference that marks
the existence of DSBs from the very early events until the restoration of the damage is
H2AX. It is widely accepted that H2AX plays a central role in the repair of DSB.
Three main pieces of evidence substantiate this conclusion: (i) the early appearance of
H2AX foci after induction of DSB, prior to all other known proteins that form foci;
(ii) the delayed disappearance of H2AX foci that is concomitant with restoration of
the lesion (iii) the abrogation of accumulation of several proteins at the lesion sites
in H2AX knockout and knockdown systems, whereas ablation of other proteins that
form foci does not have a similar impact. As mentioned in Section 2.1.3, 53BP1 and
MRN form minute foci in H2AX knockout cells, where further accumulation of these
molecules is abrogated.This evidence poses a theoretical question: how sensors such as
53BP1 and MRN are unable to recognize their primary signal in chromatin structures
all along the damaged chromatin domains in H2AX knockout cells?
It is reasonable to speculate that H2AX may recruit other factors that facilitate
chromatin changes along the broken fiber that in turn would expose methylated sites
for 53BP1 to accumulate or perhaps other structures relevant to MRN recognition.
Candidates for this role have recently emerged. Chromatin-remodeling subunits, known
to be otherwise involved in transcription, have been shown to be recruited by the
-phosphorylated chromatin during the DSB repair process. It remains to be experimentally tested whether they indeed play a role is in this context or not.
The overwhelming evidence for the critical function of H2AX in the DSB response,
do not necessarily require acceptance of a “hierarchical” model that opepates in signal
transduction pathways that are initiated by chromatin lesions. The fact that 53BP1
recognizes a different type of histone modification, that of methylated H3 Lys 79, and
initiates a different branch in the DSB damage response, strongly suggests that the
chromatin-initiated network of interactions is “relational” rather than “hierarchical”.
The existence of an interplay between methylation and 53BP1 from one end and
H2AX and MDC1/MRN from the other supports this notion. It is expected that in
the near future, additional factors and interactions will be identified, providing an
improved understanding of the role that H2AX, 53BP1, MDC1 and MRN may play
in this highly sophisticated network of DSB damage response.
260
Part III / DNA Damage Response, Signaling Pathways, and Tumorigenesis
REFERENCES
1. van den Bosch M, Lohman PH, Pastink A. DNA double-strand break repair by homologous recombination. Biol Chem 2002;383(6):873–92.
2. Rothkamm K, Kruger I, Thompson LH, Lobrich M. Pathways of DNA double-strand break repair
during the mammalian cell cycle. Mol Cell Biol 2003;23(16):5706–15.
3. Weterings E, van Gent DC. The mechanism of non-homologous end-joining: a synopsis of synapsis.
DNA Repair (Amst) 2004;3(11):1425–35.
4. Lees-Miller SP, Meek K. Repair of DNA double strand breaks by non-homologous end joining.
Biochimie 2003;85(11):1161–73.
5. Kaina B. DNA damage-triggered apoptosis: critical role of DNA repair, double-strand breaks, cell
proliferation and signaling. Biochem Pharmacol 2003;66(8):1547–54.
6. Jackson SP. Sensing and repairing DNA double-strand breaks. Carcinogenesis 2002;23(5):687–96.
7. Rogakou EP, Boon C, Redon C, Bonner WM. Megabase chromatin domains involved in DNA
double-strand breaks in vivo. J Cell Biol 1999;146(5):905–16.
8. Rogakou EP, Pilch DR, Orr AH, Ivanova VS, Bonner WM. DNA double-stranded breaks induce
histone H2AX phosphorylation on serine 139. J Biol Chem 1998;273(10):5858–68.
9. Redon C, Pilch D, Rogakou E, Sedelnikova O, Newrock K, Bonner W. Histone H2A variants H2AX
and H2AZ. Curr Opin Genet Dev 2002;12(2):162–9.
10. Shroff R, Arbel-Eden A, Pilch D, Ira G, Bonner WM, Petrini JH, Haber JE, Lichten M. Distribution
and dynamics of chromatin modification induced by a defined DNA double-strand break. Curr Biol
2004;14(19):1703–11.
11. Unal E, Arbel-Eden A, Sattler U, et al. DNA damage response pathway uses histone modification
to assemble a double-strand break-specific cohesin domain. Mol Cell 2004;16(6):991–1002.
12. Bassing CH, Alt FW. The cellular response to general and programmed DNA double strand breaks.
DNA Repair (Amst) 2004;3(8–9):781–96.
13. Bassing CH, Suh H, Ferguson DO, et al. Histone H2AX: a dosage-dependent suppressor of oncogenic
translocations and tumors. Cell 2003;114(3):359–70.
14. Chen HT, Bhandoola A, Difilippantonio MJ, et al. Response to RAG-mediated VDJ cleavage by
NBS1 and gamma-H2AX. Science 2000;290(5498):1962–5.
15. Nakamura TM, Du LL, Redon C, Russell P. Histone H2A phosphorylation controls Crb2 recruitment
at DNA breaks, maintains checkpoint arrest, and influences DNA repair in fission yeast. Mol Cell
Biol 2004;24(14):6215–30.
16. Downs JA, Lowndes NF, Jackson SP. A role for Saccharomyces cerevisiae histone H2A in DNA
repair. Nature 2000;408(6815):1001–4.
17. Burma S, Chen BP, Murphy M, Kurimasa A, Chen DJ. ATM phosphorylates histone H2AX in
response to DNA double-strand breaks. J Biol Chem 2001;276(45):42462–7.
18. Shiloh Y. ATM and related protein kinases: safeguarding genome integrity. Nat Rev Cancer
2003;3(3):155–68.
19. Riballo E, Kuhne M, Rief N, et al. A pathway of double-strand break rejoining dependent upon
ATM, Artemis, and proteins locating to gamma-H2AX foci. Mol Cell 2004;16(5):715–24.
20. Stiff T, O’Driscoll M, Rief N, Iwabuchi K, Lobrich M, Jeggo PA. ATM and DNA-PK function redundantly to phosphorylate H2AX after exposure to ionizing radiation. Cancer Res 2004;64(7):2390–6.
21. Andegeko Y, Moyal L, Mittelman L, Tsarfaty I, Shiloh Y, Rotman G. Nuclear retention of ATM at
sites of DNA double strand breaks. J Biol Chem 2001;276(41):38224–30.
22. Hammond EM, Dorie MJ, Giaccia AJ. ATR/ATM targets are phosphorylated by ATR in response
to hypoxia and ATM in response to reoxygenation. J Biol Chem 2003;278(14):12207–13.
23. Ward IM, Chen J. Histone H2AX is phosphorylated in an ATR-dependent manner in response to
replicational stress. J Biol Chem 2001;276(51):47759–62.
24. Furuta T, Takemura H, Liao ZY, et al. Phosphorylation of histone H2AX and activation of Mre11,
Rad50, and Nbs1 in response to replication-dependent DNA double-strand breaks induced by
mammalian DNA topoisomerase I cleavage complexes. J Biol Chem 2003;278(22):20303–12.
25. Koundrioukoff S, Polo S, Almouzni G. Interplay between chromatin and cell cycle checkpoints in
the context of ATR/ATM-dependent checkpoints. DNA Repair (Amst) 2004;3(8–9):969–78.
26. Fernandez-Capetillo O, Lee A, Nussenzweig M, Nussenzweig A. H2AX: the histone guardian of the
genome. DNA Repair (Amst) 2004;3(8–9):959–67.
Chapter 13 / Interplay Between H2AX and 53BP1
261
27. Paull TT, Rogakou EP, Yamazaki V, Kirchgessner CU, Gellert M, Bonner WM. A critical role
for histone H2AX in recruitment of repair factors to nuclear foci after DNA damage. Curr Biol
2000;10(15):886–95.
28. Chan DW, Chen BP, Prithivirajsingh S, et al. Autophosphorylation of the DNA-dependent protein
kinase catalytic subunit is required for rejoining of DNA double-strand breaks. Genes Dev
2002;16(18):2333–8.
29. Siino JS, Nazarov IB, Svetlova MP, et al. Photobleaching of GFP-labeled H2AX in chromatin: H2AX
has low diffusional mobility in the nucleus. Biochem Biophys Res Commun 2002;297(5):1318–23.
30. Sedelnikova OA, Pilch DR, Redon C, Bonner WM. Histone H2AX in DNA damage and repair.
Cancer Biol Ther 2003;2(3):233–5.
31. Pilch DR, Sedelnikova OA, Redon C, Celeste A, Nussenzweig A, Bonner WM. Characteristics of
gamma-H2AX foci at DNA double-strand breaks sites. Biochem Cell Biol 2003;81(3):123–9.
32. Fernandez-Capetillo O, Celeste A, Nussenzweig A. Focusing on foci: H2AX and the recruitment of
DNA-damage response factors. Cell Cycle 2003;2(5):426–7.
33. Celeste A, Fernandez-Capetillo O, Kruhlak MJ, et al. Histone H2AX phosphorylation is dispensable
for the initial recognition of DNA breaks. Nat Cell Biol 2003;5(7):675–9.
34. Celeste A, Difilippantonio S, Difilippantonio MJ, et al. H2AX haploinsufficiency modifies genomic
stability and tumor susceptibility. Cell 2003;114(3):371–83.
35. Goldberg M, Stucki M, Falck J, et al. MDC1 is required for the intra-S-phase DNA damage
checkpoint. Nature 2003;421(6926):952–6.
36. Stewart GS, Wang B, Bignell CR, Taylor AM, Elledge SJ. MDC1 is a mediator of the mammalian
DNA damage checkpoint. Nature 2003;421(6926):961–6.
37. Stucki M, Jackson SP. MDC1/NFBD1: a key regulator of the DNA damage response in higher
eukaryotes. DNA Repair (Amst) 2004;3(8–9):953–7.
38. Xu X, Stern DF. NFBD1/MDC1 regulates ionizing radiation-induced focus formation by DNA
checkpoint signaling and repair factors. FASEB J 2003;17(13):1842–8.
39. Peng A, Chen PL. NFBD1, like 53BP1, is an early and redundant transducer mediating Chk2
phosphorylation in response to DNA damage. J Biol Chem 2003;278(11):8873–6.
40. Lukas C, Melander F, Stucki M, et al. Mdc1 couples DNA double-strand break recognition by Nbs1
with its H2AX-dependent chromatin retention. EMBO J 2004;23(13):2674–83.
41. Mochan TA, Venere M, DiTullio RA Jr, Halazonetis TD. 53BP1 and NFBD1/MDC1-Nbs1 function
in parallel interacting pathways activating ataxia-telangiectasia mutated (ATM) in response to DNA
damage. Cancer Res 2003;63(24):8586–91.
42. Shang YL, Bodero AJ, Chen PL. NFBD1, a novel nuclear protein with signature motifs of FHA
and BRCT, and an internal 41-amino acid repeat sequence, is an early participant in DNA damage
response. J Biol Chem 2003;278(8):6323–9.
43. Manke IA, Lowery DM, Nguyen A, Yaffe MB. BRCT repeats as phosphopeptide-binding modules
involved in protein targeting. Science 2003;302(5645):636–9.
44. Lee MS, Edwards RA, Thede GL, Glover JN. Structure of the BRCT repeat domain of MDC1 and
its specificity for the free COOH-terminal end of the {gamma}-H2AX histone tail. J Biol Chem
2005;280(37):32053–6.
45. Peng A, Chen PL. NFBD1/Mdc1 mediates ATR-dependent DNA damage response. Cancer Res
2005;65(4):1158–63.
46. Lee AC, Fernandez-Capetillo O, Pisupati V, Jackson SP, Nussenzweig A. Specific association of
mouse MDC1/NFBD1 with NBS1 at sites of DNA-damage. Cell Cycle 2005;4(1):177–82.
47. Bekker-Jensen S, Lukas C, Melander F, Bartek J, Lukas J. Dynamic assembly and sustained
retention of 53BP1 at the sites of DNA damage are controlled by Mdc1/NFBD1. J Cell Biol
2005;170(2):201–11.
48. D’Amours D, Jackson SP. The Mre11 complex: at the crossroads of DNA repair and checkpoint
signalling. Nat Rev Mol Cell Biol 2002;3(5):317–27.
49. Petrini JH, Stracker TH. The cellular response to DNA double-strand breaks: defining the sensors
and mediators. Trends Cell Biol 2003;13(9):458–62.
50. Costanzo V, Robertson K, Bibikova M, et al. Mre11 protein complex prevents double-strand break
accumulation during chromosomal DNA replication. Mol Cell 2001;8(1):137–47.
51. Moreno-Herrero F, de Jager M, Dekker NH, Kanaar R, Wyman C, Dekker C. Mesoscale conformational changes in the DNA-repair complex Rad50/Mre11/Nbs1 upon binding DNA. Nature
2005;437(7057):440–3.
262
Part III / DNA Damage Response, Signaling Pathways, and Tumorigenesis
52. Williams RS, Tainer JA. A nanomachine for making ends meet: MRN is a flexing scaffold for the
repair of DNA double-strand breaks. Mol Cell 2005;19(6):724–6.
53. Kobayashi J, Antoccia A, Tauchi H, Matsuura S, Komatsu K. NBS1 and its functional role in the
DNA damage response. DNA Repair (Amst) 2004;3(8–9):855–61.
54. Kobayashi J, Tauchi H, Sakamoto S, et al. NBS1 localizes to gamma-H2AX foci through interaction
with the FHA/BRCT domain. Curr Biol 2002;12(21):1846–51.
55. Kang J, Ferguson D, Song H, et al. Functional Interaction of H2AX, NBS1, and p53 in ATMdependent DNA damage responses and tumor suppression. Mol Cell Biol 2005;25(2):661–70.
56. Horejsi Z, Falck J, Bakkenist CJ, Kastan MB, Lukas J, Bartek J. Distinct functional domains of
Nbs1 modulate the timing and magnitude of ATM activation after low doses of ionizing radiation.
Oncogene 2004;23(17):3122–7.
57. Difilippantonio S, Celeste A, Fernandez-Capetillo O, et al. Role of Nbs1 in the activation of the Atm
kinase revealed in humanized mouse models. Nat Cell Biol 2005;7(7):675–85.
58. Lee JH, Paull TT. Direct activation of the ATM protein kinase by the Mre11/Rad50/Nbs1 complex.
Science 2004;304(5667):93–6.
59. Uziel T, Lerenthal Y, Moyal L, Andegeko Y, Mittelman L, Shiloh Y. Requirement of the MRN
complex for ATM activation by DNA damage. EMBO J 2003;22(20):5612–21.
60. Girard PM, Riballo E, Begg AC, Waugh A, Jeggo PA. Nbs1 promotes ATM dependent phosphorylation events including those required for G1/S arrest. Oncogene 2002;21(27):4191–9.
61. You Z, Chahwan C, Bailis J, Hunter T, Russell P. ATM activation and its recruitment to damaged
DNA require binding to the C terminus of Nbs1. Mol Cell Biol 2005;25(13):5363–79.
62. Lee AC, Fernandez-Capetillo O, Pisupati V, Jackson SP, Nussenzweig A. Specific association of
mouse MDC1/NFBD1 with NBS1 at sites of DNA-damage. Cell Cycle 2005;4(1):177–82.
63. Iwabuchi K, Bartel PL, Li B, Marraccino R, Fields S. Two cellular proteins that bind to wild-type
but not mutant p53. Proc Natl Acad Sci USA 1994;91(13):6098–102.
64. Schultz LB, Chehab NH, Malikzay A, Halazonetis TD. p53 binding protein 1 (53BP1) is an early
participant in the cellular response to DNA double-strand breaks. J Cell Biol 2000;151(7):1381–90.
65. Anderson L, Henderson C, Adachi Y. Phosphorylation and rapid relocalization of 53BP1 to nuclear
foci upon DNA damage. Mol Cell Biol 2001;21(5):1719–29.
66. Rappold I, Iwabuchi K, Date T, Chen J. Tumor suppressor p53 binding protein 1 (53BP1) is involved
in DNA damage-signaling pathways. J Cell Biol 2001;153(3):613–20.
67. Sengupta S, Robles AI, Linke SP, et al. Functional interaction between BLM helicase and 53BP1 in
a Chk1-mediated pathway during S-phase arrest. J Cell Biol 2004;166(6):801–13.
68. Richie CT, Peterson C, Lu T, Hittelman WN, Carpenter PB, Legerski RJ. hSnm1 colocalizes and
physically associates with 53BP1 before and after DNA damage. Mol Cell Biol 2002;22(24):8635–47.
69. Ward IM, Minn K, Jorda KG, Chen J. Accumulation of checkpoint protein 53BP1 at DNA breaks
involves its binding to phosphorylated histone H2AX. J Biol Chem 2003;278(22):19579–82.
70. Kao GD, McKenna WG, Guenther MG, Muschel RJ, Lazar MA, Yen TJ. Histone deacetylase 4
interacts with 53BP1 to mediate the DNA damage response. J Cell Biol 2003;160(7):1017–27.
71. Mochan TA, Venere M, DiTullio RA Jr, Halazonetis TD. 53BP1, an activator of ATM in response
to DNA damage. DNA Repair (Amst) 2004;3(8–9):945–52.
72. Huyen Y, Zgheib O, Ditullio RA Jr, et al. Methylated lysine 79 of histone H3 targets 53BP1 to DNA
double-strand breaks. Nature 2004;432(7015):406–11.
73. Joo WS, Jeffrey PD, Cantor SB, Finnin MS, Livingston DM, Pavletich NP. Structure of the
53BP1 BRCT region bound to p53 and its comparison to the Brca1 BRCT structure. Genes Dev
2002;16(5):583–93.
74. Derbyshire DJ, Basu BP, Serpell LC, et al. Crystal structure of human 53BP1 BRCT domains bound
to p53 tumour suppressor. EMBO J 2002;21(14):3863–72.
75. Lo KW, Kan HM, Chan LN, et al. The 8 kDa dynein light chain binds to 53BP1 and mediates DNA
damage-induced p53 nuclear accumulation. J Biol Chem 2005;280(9):8172–9.
76. Charier G, Couprie J, Alpha-Bazin B, et al. The Tudor tandem of 53BP1: a new structural motif
involved in DNA and RG-rich peptide binding. Structure (Camb) 2004;12(9):1551–62.
77. Iwabuchi K, Basu BP, Kysela B, et al. Potential role for 53BP1 in DNA end-joining repair through
direct interaction with DNA. J Biol Chem 2003;278(38):36487–95.
78. Fernandez-Capetillo O, Chen HT, Celeste A, et al. DNA damage-induced G2-M checkpoint activation
by histone H2AX and 53BP1. Nat Cell Biol 2002;4(12):993–7.
Chapter 13 / Interplay Between H2AX and 53BP1
263
79. Jullien D, Vagnarelli P, Earnshaw WC, Adachi Y. Kinetochore localisation of the DNA damage
response component 53BP1 during mitosis. J Cell Sci 2002;115(Pt 1):71–9.
80. Abraham RT. Checkpoint signalling: focusing on 53BP1. Nat Cell Biol 2002;4(12):E277–9.
81. Wang B, Matsuoka S, Carpenter PB, Elledge SJ. 53BP1, a mediator of the DNA damage checkpoint.
Science 2002;298(5597):1435–8.
82. DiTullio RA Jr, Mochan TA, Venere M, et al. 53BP1 functions in an ATM-dependent checkpoint
pathway that is constitutively activated in human cancer. Nat Cell Biol 2002;4(12):998–1002.
83. Wysocki R, Javaheri A, Allard S, Sha F, Cote J, Kron SJ. Role of dot1-dependent histone
h3 methylation in g1 and s phase DNA damage checkpoint functions of rad9. Mol Cell Biol
2005;25(19):8430–43.
84. Sanders SL, Portoso M, Mata J, Bahler J, Allshire RC, Kouzarides T. Methylation of histone H4
lysine 20 controls recruitment of Crb2 to sites of DNA damage. Cell 2004;119(5):603–14.
85. Onclercq-Delic R, Calsou P, Delteil C, Salles B, Papadopoulo D, Amor-Gueret M. Possible antirecombinogenic role of Bloom’s syndrome helicase in double-strand break processing. Nucleic Acids
Res 2003;31(21):6272–82.
86. Celeste A, Petersen S, Romanienko PJ, et al. Genomic instability in mice lacking histone H2AX.
Science 2002;296(5569):922–7.
87. Zgheib O, Huyen Y, Ditullio RA Jr, et al. ATM signaling and 53BP1. Radiother Oncol
2005;76(2):119–22.
88. Bakkenist CJ, Kastan MB. DNA damage activates ATM through intermolecular autophosphorylation
and dimer dissociation. Nature 2003;421(6922):499–506n.